"Antisense" technology encompasses the use of sequence-specific oligonucleotides (ODNs) to bind to mRNA and inhibit its function. A variation of the "antisense" approach to rational drug design is termed "anti-gene". Whereas antisense ODNs target single stranded mRNA, anti-gene ODNs hybridize with and are capable of inhibiting the function of double-stranded DNA. More specifically, anti-gene ODNs form sequence-specific triple-stranded complexes with a double stranded DNA target and thus may affect the replication or transcription of selected target genes. As is known, except for certain RNA viruses and nucleic acid-free viroids (and possibly prions), DNA is the repository for all genetic information, including regulatory control sequences and non-expressed genes, such as dormant proviral DNA genomes. In contrast, the target for antisense ODNs, mRNA, represents a very small subset of the information encoded in DNA. Furthermore, for a given gene, the mRNA transcript is present at a much higher effective concentration than the DNA which encodes the transcript. Anti-gene ODNs are, therefore, effective in much lower concentrations than antisense ODNs. Thus, anti-gene ODNs have broader applicability and are potentially more powerful than antisense ODNs that merely inhibit mRNA processing and translation.
Anti-gene ODNs in the nuclei of living cells can form sequence-specific complexes with chromosomal DNA. The resultant triplexes have been shown to inhibit restriction and/or transcription of the target double stranded DNA. Based on the known stabilities of the two target nucleic acid species (i.e., DNA and RNA), anti-gene interference with DNA functioning has longer lasting effects than the corresponding antisense inhibition of mRNA function.
As noted above, anti-gene therapy is based on the observation that under certain conditions DNA can form triple-stranded complexes. Several forms of triple-strand complex are known. In one type of triple-stranded complex, the third strand resides in the major groove of the Watson-Crick base paired double helix, where its hydrogen bonds to one of the two parental strands. A binding code governs the recognition of base pairs by a third base (see allowed triplets below, Hoogsteen or reverse Hoogsteen pairing). In each case, the third strand base is presented first (in boldface) and is followed by the base pair in the Watson-Crick duplex. allowed triplets: A-A-T G-G-C T-A-T C-G-C
Certain limitations of this base pair recognition code are apparent from the allowed triplets. First, there is no capability for the recognition of T-A and C-G base pairs; hence, triple strand formation is restricted to runs of purine bases on one strand and pyrimidine bases on the other strand of the duplex. In other words, the third strand or ODN binds only to one strand of the duplex and can only bind to purines. Second, if cytosine ("C") is in the third strand, it must be protonated to be able to hydrogen bond to the guanine of a G-C base pair. The pK.sub.a for protonation of cytosine is 4.6, suggesting that at physiological pH the stability of C-G-C triads is likely to be impaired. Third, in all cases triads are maintained by two hydrogen bonds between the third strand base and the purine residue of the duplex base pair. Hence, triple-stranded complexes are generally less stable than the parental double-stranded DNA, which is maintained by a combination of two (A-T) or three (G-C) hydrogen bonds between purine and pyrimidine pairs. (Watson-Crick motif).
An important disadvantage of triple strand formation as discussed above is the relatively slow kinetics of triple strand formation. However, triple strand formation can be catalyzed in cells by recombinase enzymes which are practically ubiquitous in cells and whose existence is well known in the art. In addition to a much faster rate of triple strand formation, recombinase enzyme-catalyzed triple strand formation also provides the advantage of universal sequence recognition (in contrast to the A-T and G-C recognition limitation associated with non-enzyme-mediated triple strand formation). More specifically, the recombinase enzyme-mediated recognition motif recognizes all four base pairs, thereby allowing targeting of any double stranded DNA sequence. Second, the nucleoprotein filament, which is the complex formed between a recombinase enzyme and the single-stranded ODN, searches for target double strand DNA homology much more efficiently than does a small naked anti-gene ODN, thus decreasing the concentration of anti-gene ODN required for efficient triple strand complex formation. Third, due to the hydrogen bonding patterns, the resultant triple strand complex is stable at physiological pH. Fourth, since cellular recombination mechanisms are being utilized, the DNA in higher order chromatin structures is accessible for targeting.
The ability to achieve targeted mutagenesis of a chromosomal gene in a living cell is a long-sought goal in the area of gene therapy. In its most effective embodiment, targeted mutagenesis would result in the change of a single nucleotide in the sequence of a chromosomal gene, for example conversion of a point mutant allele into its wild-type counterpart, or inactivation of a deleterious gene by creating a nucleotide triplet specifying translational termination. In order to direct mutagenizing agents to a particular target nucleotide sequence, a targeting mechanism having high specificity is required. Such specificity can be obtained, for a polynucleotide target, by the use of a synthetic oligonucleotide having a sequence that allows the oligonucleotide to bind to the target sequence.
Binding of an oligonucleotide to a single-stranded target can be accomplished by designing the sequence of the oligonucleotide such that it base-pairs with its target, in the Watson-Crick sense. However, it is also possible to design an oligonucleotide with a sequence that will allow it to base-pair with a duplex nucleic acid, forming a triple-stranded nucleic acid, or triplex. Fresco, U.S. Pat. No. 5,422,251. Attachment of a suitable modifying agent to such an oligonucleotide would make it possible to generate a lesion at or near a target sequence in a gene of interest. Subsequent cellular processes related to DNA replication and/or repair can result in either restoration of the original sequence by repair of the lesion, or mutagenesis, for example by misrepair, resulting in a base change at the site of the lesion.
Despite numerous attempts to develop appropriate reagents and techniques, generation of a specific mutation in a chromosomal gene by a modified oligonucleotide has yet to be demonstrated. Indeed, consideration of the possible consequences of targeted damage to the genome of a cell by a modified oligonucleotide suggests that more likely outcomes of such damage would be either cell death or regeneration of the original sequence. For instance, the presence of a low level of a specific type of damage at a particular site (as can occur when mutation is targeted to a specific site, for example, by the use of an oligonucleotide) might be expected to be readily corrected by cellular repair mechanisms. On the other hand, widespread damage at multiple sites might be anticipated to overwhelm the repair capacity of the cell, resulting in the accumulation of multiple mutations in a variety of genes and consequent cell death.
Attempts to exploit the specificity of RNA-DNA base-pairing to direct the targeting of mutagenic agents in vitro were conducted by Salganik et al. (1980) Proc. Natl. Acad. Sci. USA 77:2796-2800. In these experiments, a heterobifunctional alkylating agent was attached to mRNA transcripts of bacteriophage T7, at random locations within the transcripts. These modified mRNAs were capable of hybridization to T7 DNA, albeit at a lower efficiency than unmodified mRNA. Following hybridization of these modified mRNAs to T7 DNA, activation of the alkylating group by reduction, and packaging of the modified DNA, a low frequency of mutations was detected in progeny phage, as measured by plating efficiency on various indicator strains of the host bacterium Escherichia coli.
Although pioneering for its time, this method did not allow efficient targeting of mutational events to a particular nucleotide in a living cell. First, it was not possible to place the mutagenic alkylating group at a specific site on the modified mRNA. Therefore, site-specific mutagenesis of the target sequence was not possible. Second, modification of the mRNA reduced its efficiency of hybridization to its target. Third, hybridization was performed in vitro under R-loop conditions, which favor RNA-DNA hybridization. Finally, after hybridization of the modified mRNA to its target, activation of the mutagenic alkylating group by sodium borohydride reduction was required. These last two factors make the method unsuitable for use in living cells.
Others have demonstrated mutagenesis of a target sequence after binding of a modified triplex-forming oligonucleotide to a target ex vivo, or in vitro followed by transfection into mammalian cells, where the target DNA was assumed to undergo replication and/or repair. Glazer, PCT publication WO 95/01364. The oligonucleotides exemplified in these experiments were modified by the attachment of a psoralen, a photoactivatible crosslinking agent. However, the utility of such photoactivatible crosslinkers for targeted mutagenesis in living cells is limited, for several reasons. First, in order to obtain crosslinking, which appears to be a necessary prerequisite for mutagenesis, the triplex formed by the modified oligonucleotide must be exposed to near ultraviolet light. Since near ultraviolet light will not penetrate the skin, psoralen-modified oligonucleotides are not useful for mutagenesis in cells of internal organs or in other regions of the body not penetrable by near ultraviolet light.
A second problem associated with attempts to use psoralen-modified oligonucleotides for targeted mutagenesis in living cells derives from the fact that formation of triplexes is an equilibrium process, such that binding of a triplex-forming oligonucleotide to its target sequence is reversible. Hence, at any given moment, a certain fraction of mutagenic oligonucleotides are bound in triplexes with target and a certain fraction remain unbound. When the photoactivatible triplex-forming oligonucleotides of the prior art are illuminated to induce crosslink formation, only those oligonucleotides bound to their target at the moment of illumination will form crosslinks which have the potential to generate a mutation. Thus, in order to obtain maximal degrees of mutagenesis, either constant irradiation of cells with near ultraviolet light, or extremely high occupancy of binding sites (that is not characteristic of the binding equilibrium between modified oligonucleotide and target sequence) would be required.
A third problem with the potential use of psoralen-modified oligonucleotides for targeted mutagenesis of living cells is that the ultraviolet irradiation required for activation of the psoralen group is likely to have nonspecific adverse effects on cellular macromolecules. For instance, ultraviolet irradiation will induce the formation of thymine dimers in DNA, which would likely lead to nonspecific generation of mutations outside of the target site. Alternatively, or in addition to its effects on DNA, ultraviolet light can cause protein crosslinking, which could lead to various types of cellular malfunction or cell death.
A fourth problem is that psoralen itself has a non-specific affinity for DNA. As a result, a high proportion of lesions are expected to occur at sites other than the target site when psoralen-containing agents are used.
Hence, there remains a great deal of uncertainty regarding the ability of modified oligonucleotides to cause specific, non-lethal nucleotide sequence changes in the genome of a living cell. To make practical the targeted mutagenesis of a gene in a living cell, it would be necessary to develop an agent which is capable of specific interaction with a target double-stranded nucleotide sequence, and which has an intrinsic ability to react with genomic DNA (i.e., does not require external activation) in the nucleus of a cell.
All patents and publications mentioned herein, either supra or infra, are hereby incorporated by reference in their entirety.